The Mitochondrial Intermembrane Space: A Hub for Oxidative Folding Linked to Protein Biogenesis

Abstract

Significance: The introduction of disulfide bonds in proteins of the mitochondrial intermembrane space (IMS) is fundamental for their folding and assembly. This oxidative folding process depends on the disulfide donor/import receptor Mia40 and the flavin adenine dinucleotide oxidase Erv1 and concerns proteins involved in mitochondrial biogenesis, respiratory complex assembly, and metal transfer. Recent Advances: The recently determined structural basis of the interaction between Mia40 and some substrates provides a framework for the electron transfer process. A possible proofreading role for the cellular reductant glutathione has been proposed, while other studies suggest the association of Mia40 and Erv1 in dynamic multiprotein complexes in the IMS. Critical Issues: The association of Mia40 with Erv1 and substrates in large multiprotein complexes is critical. Completion of substrate folding by additional disulfide bonds after initial binding to Mia40 remains unclear. Furthermore, a more general role for Mia40 in recognizing substrates targeted to other compartments, or even without specific cysteine motifs, remains an intriguing possibility. Future Directions: Dissecting a regulatory role of intramitochondrial protein complex organization and small redox-active molecules will be crucial for understanding oxidative folding in the IMS. This should have an impact on the physiology of human cells, as disease-linked mutations of key components of this process have been manifested, and their expression in stem cells appears crucial for development. Antioxid. Redox Signal. 19, 54–62.

Introduction

M itochondrial biogenesis relies on dedicated protein targeting/import pathways, as most mitochondrial proteins are synthesized in the cytosol and subsequently translocated to one of the different mitochondrial subcompartments (Fig. 1). Import is facilitated by the translocase of the outer mitochondrial membrane (TOM) complex, which functions as the general import pore, and the TIM23 complex that targets proteins to the matrix. Integration in the membranes requires either the sorting and assembly machinery complex in the outer membrane or the TIM22 complex in the inner membrane (12).

FIG. 1.

Schematic representation of all the import pathways in mitochondria. All precursors engage first at the translocase of the outer mitochondrial membrane (TOM) import channel and subsequently diverge following distinct pathways. (A) The mitochondrial import and assembly (MIA) pathway for the oxidative folding of intermembrane space (IMS) proteins containing cysteine residues shown as dots; (B) the stop-transfer pathway used by soluble IMS proteins that pass through the TIM23 channel first and then diffuse into the lipid bilayer. Cleavage of the stop-transfer signal by an IMS-resident protease releases the soluble protein into the IMS, (C) the matrix pathway: preproteins enter through the TIM23 channel, are drawn by the action of the membrane potential (Δψ) and the translocation motor mtHSP70, and fold into an ordered structure after cleavage of the presequence by the matrix-resident mitochondrial processing peptidase (MPP); (D) the carrier pathway used by proteins destined for the inner membrane: the preproteins interact with the small Tim chaperones in the IMS and are then targeted to the inner membrane by the TIM22 complex.

Import into the intermembrane space (IMS) follows different pathways. The oxidative folding pathway has received a lot of attention recently, although it was initially seen with skepticism, because of the widespread belief that reduced glutathione (GSH) can freely diffuse into the IMS rendering it a reducing environment. Contrary to common belief, the concept of oxidative protein folding in the IMS was proposed based on the biogenesis of the cysteine-rich small Tims and three main observations: (i) correct folding of small Tims depends on intramolecular disulfides formed in vivo; (ii) oxidation of small Tims before import inhibits their import; and (iii) oxidation in vitro was too slow, suggesting that a protein-catalyzed process must occur in vivo (23). This key protein turned out to be Mia40, which was discovered in studies of proteins of unknown function identified by a comprehensive mass spectrometry analysis of the mitochondrial Saccharomyces cerevisiae proteome (13).

The MIA machinery controls directly the oxidative folding of IMS proteins, thereby trapping them in a folded and functional state (13, 25). In the first step, the disulfide donor protein Mia40 has the dual function of (i) specifically recognizing the imported precursors acting as an IMS receptor and (ii) introducing into them disulfide bonds, thereby crucially affecting their folding. The Mia40-catalyzed reaction results in reduction of its redox-active center, which is reoxidized in a second reaction that requires the flavin adenine dinucleotide (FAD)-linked sulfhydryl oxidase, Erv1 (21, 25). Completion of the electron transfer process subsequently involves cytochrome C or cytochrome C peroxidase, the cytochrome oxidase complex, and molecular oxygen (2, 14).

Recent studies of this pathway have focused on the interaction between Mia40 and the substrates and the interaction between Mia40 and Erv1 using a combination of efficient reconstitution systems, import assays, and structural analyses. In the present review, we will discuss current progress in the identification of substrates for this pathway, the salient features of the two key proteins Mia40 and Erv1, our current understanding of the basic mechanism, and recent issues on the molecular and spatial control of this process. Finally, we will refer to the relevance of mitochondrial oxidative folding for human cells and possible links to disease.

The Substrates: How Many and How Different?

Typical substrates of the MIA pathway are preproteins with conserved CX3C motifs, such as the well-studied examples of the small Tims. These function as multiprotein chaperones that facilitate passage of membrane proteins across the IMS to either the inner or the outer membrane (12). In addition, the IMS hosts a wide range of proteins with duplicated CX9C motifs, such as the copper chaperones Cox17, Cox19, and Cox23, which contribute to the assembly of cytochrome C oxidase (9), and also the Mdm35 protein that is involved in import of factors controlling lipid homeostasis in mitochondria (29). All of the above examples are proteins with twin CXnC motifs (n=3 or 9) adopting a coiled coil–helix-coiled coil–helix (CHCH) fold. A detailed search of the Saccharomyces cerevisiae database for proteins containing a twin CX9C motif revealed at least 59 such proteins (15), from which a total of 10 were predicted to have a CHCH fold and shown to localize to mitochondria. In an independent approach, Longen et al. (22) identified four additional members of the twin CX9C protein family in mitochondria, raising this complement to 14 different polypeptides. Most of these proteins are involved in the assembly and/or stability of the mitochondrial respiratory chain complexes. The twin CX9C family is conserved in all eukaryotes, with several hundred family members in eukaryotic cells (11). On the other hand, exhaustive searches of the Saccharomyces cerevisiae and other eukaryotic genomes for proteins containing the twin CX3C motif converge to the 5 members of the small Tim family (Tim 8, 9, 10, 12, and 13) as the only ones in this group (22). Interesting deviations from the classical CHCH substrates have been reported recently (Table 1). This is the case of Ccs1, which is required for the oxidation and activation of superoxide dismutase 1 (30). Ccs1 is a 27-kDa protein that differs from the CHCH substrates, as it has three distinct domains and forms only one disulfide by the action of Mia40 (18). Another protein, the iron–sulfur cluster protein Dre2, was additionally shown to form mixed-disulfide intermediates with Mia40 in vitro (6), although in vivo evidence that Dre2 is oxidized and imported by Mia40 is still lacking. Dre2 is an early component of the cytosolic iron–sulfur cluster assembly machinery in the cytosol (28), but was also found in small amounts in the IMS of yeast mitochondria (42). This protein has a C-terminal CX2C segment where Mia40 apparently binds. This expands the repertoire of Mia40 substrates, as it is the first case of an Fe/S cluster protein that interacts with Mia40. The main components of the MIA pathway themselves (i.e., Mia40 and Erv1) are substrates of Mia40 (35). Since Mia40 has a CX9C structural motif, one can draw parallels with other CHCH substrates to the way this may interact with Mia40, although this is not yet known. Erv1 on the other hand bears no structural similarity and has no sequence homology to any other Mia40 substrate. It is therefore very intriguing to understand how it may interact with Mia40. It would be of interest to understand how these two components fully assemble into their active state. Mechanistic and structural studies revealed the way Mia40 recognizes some of its substrates, allowing us to build a framework that can provide a common basis for the action of Mia40.

Table 1.

Classification of the Different Substrates for the Mitochondrial Import and Assembly Pathway

Yeast protein	Cysteine motif present	Function
Small Tims	Double CX₃C	Chaperone
Cox family	Double CX₃C	Copper chaperone
Erv1	CX₂C…CX₂C…CX₁₆C	Oxidative folding in the IMS and Fe/S cluster biogenesis
Mia40	Double CX₉C	Oxidative folding in the IMS
Ccs1	…CX₂CX₆CX₃₆CX_nC…	Copper chaperone
Dre2	Double CX₂C	Fe/S cluster biogenesis
Mdm35	Double CX₉C	Mitochondrial morphology
Mic17, Mic14	Double CX₉C	Mitochondrial respiration

The CHCH proteins (with either CX3C or CX9C motifs as indicated) are the ones studied in greater detail, while an expanding number of other substrates for Mia40 with more complex structures has been identified.

The Key Players: Mia40 and Erv1

The IMS receptor Mia40

To date, most studies focus on the Mia40 role as the key IMS oxidoreductase (5, 13, 25). In yeast, Mia40 gets anchored at the inner membrane via its N-terminus (Fig. 2) using TIM23, leaving the catalytic C-terminal domain exposed to the IMS (13). This 44-kDa protein contains a conserved cysteine-proline-cysteine tripeptide (CPC) redox active site (5, 17, 36) with the second cysteine being the catalytic one (5). This cysteine makes a mixed disulfide intermediate with the preprotein (5, 17, 36). Solution nuclear magnetic resonance (NMR) analysis reported the basic structural features of the human Mia40 (5), and was supported by additional studies on the crystal structure of the soluble part of yeast Mia40 (17). These studies revealed a structural core stabilized by two intramolecular disulfide bonds between the twin CX9C motifs, which do not participate in the oxidation reaction with the substrate (5). A characteristic feature of the core is the presence of a rather shallow, open cleft, made of several hydrophobic residues and proposed to serve as a substrate-binding site. This concept was experimentally proven by mutagenesis combined with in vitro assays and yeast complementation (5, 17). The CPC motif sits on top of the substrate-binding cleft, ideally suited to make the transient disulfide with the substrate once the latter has been positioned favorably onto the cleft of Mia40 (5).

FIG. 2.

The disulfide relay and electron flow reactions underpinning oxidative folding in the mitochondrial IMS. Schematic representation of the electron flow in yeast and mammalian cells is indicated. Mia40 is membrane-anchored only in yeast cells. Mammalian Erv1 (called augmenter of liver regeneration [ALR]) is a covalent head-to-tail dimer (dimer-specific disulfides shown in dashed lines), whereas yeast Erv1 is a noncovalent dimer. The electron flow path follows from the substrate to the cysteine-proline-cysteine tripeptide (CPC) motif of Mia40, to the N-terminal shuttle of Erv1, to the flavin adenine dinucleotide (FAD)-proximal disulfide, to FAD, to CytC, and finally to cytochrome C oxidase (COX) and oxygen. The interaction between mammalian ALR and CytC is based on in vitro studies, while in vivo evidence is not yet available.

Sulfhydryl oxidase, Erv1

The IMS also harbors the essential FAD-binding protein Erv1 (member of the Erv1/quiescin–sulfhydryl oxidase family) (21) (Fig. 2). The X-ray structure of the rat homolog reveals that FAD is coordinated noncovalently at the C-terminal by a four-helix bundle (40). Yeast Erv1 has three conserved cysteine pairs, C30/C33 at the N-terminus, C130/C133 proximal to the FAD moiety, and C159/C176 at the carboxyl end. The FAD-proximal cysteine pair C130/C133 (proximal disulfide) is thought to be involved in the electron transfer, while the C159/C176 pair has only a structural role (structural disulfide) without a direct involvement in the electron transfer process. The C30/C33 pair (shuttle disulfide) is localized at the N-terminus of the protein, which is thought to be unfolded, and participates in the interaction with Mia40 (see below). Several studies have indicated that Erv1 forms a noncovalent dimer by contacts between the core FAD domains of the two protomers; the flexible N-terminal end from one protomer is proposed to interact with the FAD-proximal disulfide of the other protomer in a defined intermolecular transfer of electrons along the Erv1 dimer (7) to the FAD moiety, and from then on to the final electron acceptors.

Apart from recycling, the catalytic center of Mia40 to its oxidized state, Erv1, was also found to play an important role in the maturation of iron–sulfur clusters as a part of the ISC export machinery. Deletion of Erv1 in yeast cells results in morphological defects of mitochondria, loss of the mitochondrial genome, and lethal phenotypes (21). It is still unclear whether the two apparently different roles of Erv1 in protein biogenesis—as a component of the MIA pathway and as a key mediator of the export of Fe/S clusters—share any links.

The Electron Cascade: The Basic Mechanism

The cytosolic targeting of Mia40 substrates remains unknown. It has been shown that for its entry via the TOM channel, these precursors must be reduced (and hence unfolded) (13, 23). Interaction with Mia40 is guided by an internal targeting signal (called intermembrane space targeting signal [ITS] (32) or mitochondrial intermembrane space sorting signal) (27), which has the tendency to form an amphipathic helix with conserved hydrophobic residues grouped in the one side of the helix where the docking cysteine is also found. This signal is 9-amino-acid-residue long and adjacent (upstream or downstream) to the cysteine that docks to the CPC motif of Mia40. Docking relies on a specific cysteine for the small Tims (26, 33) and Cox17 (32). The ITS could even target nonmitochondrial proteins to the organelle displaying thus an independent targeting capacity (32). Interestingly, the ITS is sufficient for translocation across the OM, but specific interactions with OM components remain to be identified (27, 32).

The initial demonstration that hydrophobic interactions guide the ITS onto the Mia40-binding cleft (32) was followed by detailed studies of isolated, stable complexes of Mia40 with either the ITS peptide or the entire substrate Cox17 (4). This represented the first such analysis at atomic resolution between an entire substrate polypeptide en route in the translocation process as a complex with its specific protein import component (Mia40 in this case). It was found that the Mia40 substrate binding occurs in two steps. In each step, an induced helical folding reaction is coupled to disulfide bond formation. In the first step, Mia40 induces the localized folding of the ITS coupled to the intermolecular disulfide pairing between the CPC of Mia40 and the substrate docking cysteine (Fig. 3). In the second step, the now partly folded ITS functions as a template to induce the folding of the second helix of the substrate coupled to the formation of the second intramolecular disulfide.

FIG. 3.

Mechanism of substrate recognition by Mia40. Step 1: Sliding of the substrate onto the Mia40 surface orients the ITS of the substrate (bold line) within the binding cleft of Mia40 via hydrophobic interactions. Step 2: Docking of the substrate is effected by covalent attachment to Mia40 via an intermolecular disulfide bond (with the second Cysteine of the active-site CPC motif of Mia40) and is coupled to induced helical folding of the intermembrane space-targeting signal (ITS) by Mia40. Step3: Release of the substrate is triggered by nucleophilic attack of the substrate-docking cysteine participating in the intermolecular bond with Mia40 by the partner cysteine of the substrate. Step 4: Maturation of the substrate is completed by folding of the second helix coupled to the formation of the second intramolecular disulfide (for coiled coil–helix-coiled coil–helix [CHCH] substrates), or by other factors (ligand binding, etc., depending on the substrate). Dashed lines indicate the disulfide pairs, and dots the cysteine residues. Details of this mechanism are explained in the text and references (2) and (3).

Mia40 binding can therefore be described as an α-helical chain reaction. The key role of Mia40 is that of a chaperone that induces folding of only the first helix (the ITS) (4). Subsequent folding steps may differ for different substrates, and may not necessarily involve Mia40, but other accessory molecules (ligands, prosthetic groups, etc.) or even depend entirely on other secondary structure elements of the substrates (such as the presence of the second helix for Cox17 and Tim10). In this first step of the oxidative folding pathway, electrons flow from the substrate to the CPC motif of Mia40 (making just one internal disulfide on the substrate). The second disulfide has been attributed to different factors, including a second molecule of Mia40 (7), but it is not yet clear how this may occur. Reoxidizing the CPC motif requires electron flow from the CPC motif to the shuttle disulfide of Erv1, driven by hydrophobic interactions and molecular mimicry (3, 20). This Mia40-Erv1 interaction is guided by a specific peptide segment in the region downstream of the shuttle motif in Erv1 (a stretch that is rich in conserved hydrophobic residues), which binds to the Mia40 cleft in a manner similar to that of the ITS (substrate mimicry) (3). NMR analyses of the transient complex between Mia40 and the augmenter of liver regeneration support this mechanism. It is worth noting that according to this mechanism, the substrate and Erv1 cannot bind concomitantly to one Mia40 molecule, as they would compete for the same hydrophobic binding cleft. Electrons flow from the shuttle motif of Erv1 to the proximal CXXC disulfide of Erv1 and from then on to FAD. Terminal electron acceptors can be molecular oxygen, cyt C, cytochrome C peroxidase (2, 14, 16), or yet unknown molecules under anaerobic conditions. The transfer of electrons is supported by thermodynamic redox potentials measured for some of the disulfide pairs of the cascade (5, 14), but a comprehensive kinetic analysis of the subreactions is still missing. Support for the process as described above has been obtained by reconstitution studies (7, 16, 38).

Finally, although metal binding does not affect the interaction of Mia40 with the substrate (33), recombinant Mia40 has been shown to bind zinc and copper, resulting in enhanced proteolytic stability of the protein (37). Moreover, the IMS protein Hot13 has been proposed to act as a metal-binding factor employed to improve the efficiency of the Mia40 pathway by removing zinc ions bound on imported polypeptides and Mia40 itself (24).

Regulation of the Basic Oxidative Folding Process

Substantial progress has been accomplished in our understanding of the mechanism, but recent findings raise the possibility that several factors may control the fine-tuning and efficiency of this process.

Ternary complex between Mia40-Erv1-substrate

Mia40 has been shown in vitro to be sufficient for the introduction of both disulfide bonds in the substrate (5, 7, 16). Different scenarios have been proposed on how the second disulfide in the CHCH substrates is introduced. One suggestion involved oxygen (5), even though this cannot explain how this would work under anaerobic conditions. An alternative speculation suggested that Mia40 performs a repeat cycle on the substrate (7), but given the specific binding of Mia40 to the ITS of the substrate (27, 32), it is unclear where Mia40 could bind on the substrate during the second binding round. Alternatively, it was proposed that complete oxidation requires Erv1 (34), a concept supported by the finding of a ternary complex between the precursor, Mia40, and Erv1 (Fig. 4). The three-component complex was proposed to remain stable, as Mia40 oxidizes both disulfides on the substrate, with Erv1 having an important, but indirect, role in this process. The existence of the ternary complex would allow the efficient spatial channeling of electrons in a physical conduit for the completion of the electron transfer process. This is an attractive hypothesis, but given that the concomitant binding of the substrate and Erv1 to Mia40 is mutually exclusive (4), further work is needed to map possible interacting surfaces and understand the molecular architecture of such a ternary complex. In the same vein, the molar excess of Mia40 over Erv1 in isolated mitochondria (4) suggests that the interaction between Mia40 and Erv1 is a dynamic one and cannot be readily explained by the occurrence of a ternary complex.

FIG. 4.

The ternary complex between the substrate, Mia40, and Erv1. It is proposed that formation of the ternary complex is necessary for efficient channeling of electrons in the oxidative folding process (34). The noncovalent dimer of Erv1 must be associated to Mia40 in a manner that allows both efficient binding and oxidation of the substrate by Mia40 and also efficient recycling of the CPC of Mia40 back to its oxidized state by the action of the N-terminal shuttle cysteine pair of Erv1.

A role for glutathione?

Is there an isomerization pathway in the mitochondrial IMS? This is important, as in all cellular compartments where oxidative folding occurs, dedicated isomerization reactions exist to improve the efficiency of the process. In the bacterial periplasm, DsbC performs isomerization reactions to promote correct disulfide bonding (41). In the endoplasmic reticulum (ER), protein disulfide isomerase (PDI) works both for the introduction and reshuffling of disulfide bonds, exhibiting thus isomerase activity (31). We should however note that the Mia40 system is much more specific than either Dsb or PDI in terms of recognition of the substrate, as none of these systems is underpinned by specific substrate-targeting sequences that orient precisely the substrate to the oxidase. By contrast to DsbA and PDI that lack any selectivity and can interact with any accessible free cysteine, Mia40 can only do so to the specific docking cysteine of the substrate that is primed for interaction with Mia40 via correct positioning within the cleft of Mia40. In this sense, a requirement for an isomerization pathway may not be as stringent for the IMS, as it is for the periplasm or the ER. Still, in a recent study, a proofreading role for the abundant cellular reductant GSH in the IMS has been proposed (7) (Fig. 5). Specifically, it was shown that upon oxidation of the substrates in vitro, Mia40 forms partially oxidized long-lived intermediates that are prevented in the presence of GSH. Moreover, glutathione was shown to increase the efficiency of precursor import in isolated mitochondria. Since one can argue that any small-molecule reductant could serve such a role in vitro, it would be worthwhile to further assess the role of GSH in controlling the oxidative folding in the IMS in intact cells and whether it is relevant to the expanding repertoire of novel in vivo Mia40 substrates or only to a subset of them. It is not known whether the GSH/oxidized glutathione ratio is tightly maintained in the mitochondrial IMS in a manner similar to the tight control of the ER lumen. It is conceivable that in addition to providing buffering capacity in the IMS as it does in the ER, GSH could also serve a role in feedback regulation of Erv1 and Mia40 as is the case for Ero1 and PDI in the ER. It is accepted that feedback regulation of Ero1 activity exerted by PDI is a central element in ER redox control (8). Data on a possible mechanism of feedback inhibition in the IMS are still lacking, but such a mechanism would provide a solution to the problem of over oxidation and reactive oxygen species generation in the IMS, mirroring the relevant mechanism discovered in the ER (8).

FIG. 5.

Proofreading role of reduced glutathione (GSH). The formation of long-lived mixed-disulfide intermediates between Mia40 and the substrates is proposed to be prevented by the proofreading activity of reduced glutathione, but the relevance of this or a dedicated reducing and isomerization pathway in vivo remains open (see text for discussion).

MINOS assembly and internal architecture of mitochondria

Recent studies have reported that mitofilin/Fcj1, a protein involved in the formation of the crista junctions, participates in a larger complex along with five more proteins (Mio10, Aim5, Aim13, Aim37, and Mio27), the so-called mitochondrial inner membrane organizing system (MINOS) (or MitOS or MICOS) complex to control crista morphology (39). What was intriguing, in addition to the clear importance of this complex for mitochondria internal architecture, is that this complex affected the import of MIA substrates, as it interacts with both the TOM channel and Mia40 (Fig. 6). The explanation suggested was that mitofilin brings into proximity the TOM complex with at least a fraction of Mia40 molecules to assist the interaction with the newly imported preproteins. This spatial organization would enhance the efficiency of the Mia40-dependent folding early on during the import process. This concept would support two important concepts, namely the efficient channeling of electrons (34) and the coupled binding/folding of the substrate (3). Future studies should address unresolved questions such as the participation of Erv1 (even transiently) in MINOS, and the relevance this may have for the proposed ternary complex of Mia40-Erv1-substrate.

FIG. 6.

The mitochondrial inner membrane-organizing system (MINOS) complex may bring Mia40 into the proximity of the TOM complex acting as a bridge. The MINOS complex interacts transiently with the TOM complex, at contact sites between the two membranes, assisting thus the translocation of precursor proteins. This MINOS–TOM interaction is indicated by a thick, double arrow. By interacting transiently also with Mia40 (indicated by a bold dotted double arrow), MINOS could bring Mia40 to the proximity of the TOM pore. This interaction could explain the effect of MINOS on the import of Mia40 substrates. Additionally, it is well established that Mia40 interacts with Erv1 (bold dotted arrow), while an interaction between Erv1 and the MINOS complex is unknown (dotted arrow with a question mark). The curved arrow on the left of the figure indicates the Mia40-dependent default oxidative folding pathway for the incoming Mia40 substrates.

Nonresolved Issues

Human Mia40 is only 15 kDa, lacking the yeast N-terminal region of 250 residues that contains the presequence and the segment that anchors it in the IM (13). Consequently, the human homolog is soluble in the IMS. On the other hand, the hMia40 is highly homologous to the C-terminal domain of yeast Mia40, and the two structures are completely superimposable (17), and the two proteins could be used interchangeably in binding studies with different substrates (5). Therefore, the key features and basic requirements for the Mia40 oxidase function are maintained from yeast to mammals. By contrast, the participation of Mia40 in membrane complexes (and the influence this may have on the efficiency or control of the process) differs in higher eukaryotes where Mia40 is not membrane bound. In some cells, there is no homolog for Mia40 (the case of trypanosomes (1), or the Mia40 gene is dispensable while the Erv1 homolog is essential (the case of Arabidopsis thaliana) (10). This would argue that although the de novo oxidase function of Erv1 is absolutely essential, the disulfide exchange role of Mia40 is not or could be taken over by other molecules. The latter could be the case in A. thaliana cells, where Erv1 lacks an N-terminal CX2C (equivalent to the yeast N-terminal shuttle), but has instead a C-terminal CX4C motif. This unusual architecture of A. thaliana Erv1 could conceivably take over the Mia40 disulfide-exchange role, as the Mia40 knockout is viable.

The recent progress on the recognition of CHCH substrates by Mia40 allows us to rationalize the structural basis of this process, although the completion of the folding process for some substrates in vivo remains to be elucidated. In parallel, the rapidly expanding repertoire of proteins that share no sequence homology or structural similarity and still are recognized by Mia40 raises the possibility for a more general role of Mia40. This can be assigned to import of cysteineless proteins into the IMS, or even cysteine-containing proteins targeted to compartments other than the IMS.

The control and fine tuning of oxidative folding may require spatial organization of key components into dynamic complexes between the two membranes. Additionally, small redox-active molecules such as GSH may have previously underestimated roles (19). Future studies will have to build on the development of efficient reconstitution systems to address the more challenging intracellular milieu of intact cells. This will allow us to appreciate the influence of different factors in the regulation of this process. Simple and tractable model systems like yeast hold promise to develop key regulatory concepts in this respect. On the other hand, work in human cells will shed light into the full integration of this process in redox homeostasis and the redox-signaling network in the cell.

Footnotes

Acknowledgments

Our research was supported by the THALIS and ARISTEIA grants (Ministry of Education of Greece). We thank members of our group for helpful discussions. We apologize to our colleagues whose primary work could not be cited due to the limited number of references allowed. An effort has been made to cite the most appropriate or more recent primary publications and to include reviews to cover the related work.

Abbreviations Used

References

Allen

, Ferguson

, Ginger

. Distinctive biochemistry in the trypanosome mitochondrial intermembrane space suggests a model for stepwise evolution of the MIA pathway for import of cysteine-rich proteins. FEBS Lett, 582:2817–2825. 2008.

Allen

, Balabanidou

, Sideris

, Lisowsky

, Tokatlidis

. Erv1 mediates the Mia40-dependent protein import pathway and provides a functional link to the respiratory chain by shuttling electrons to cytochrome c. J Mol Biol, 353:937–944. 2005.

Banci

, Bertini

, Calderone

, Cefaro

, Ciofi-Baffoni

, Gallo

, Kallergi

, Lionaki

, Pozidis

, Tokatlidis

. Molecular recognition and substrate mimicry drive the electron-transfer process between MIA40 and ALR. Proc Natl Acad Sci U S A, 108:4811–4816. 2011.

Banci

, Bertini

, Cefaro

, Cenacchi

, Ciofi-Baffoni

, Felli

, Gallo

, Gonnelli

, Luchinat

, Sideris

, Tokatlidis

. Molecular chaperone function of Mia40 triggers consecutive induced folding steps of the substrate in mitochondrial protein import. Proc Natl Acad Sci U S A, 107:20190–20195. 2010.

Banci

, Bertini

, Cefaro

, Ciofi-Baffoni

, Gallo

, Martinelli

, Sideris

, Katrakili

, Tokatlidis

. MIA40 is an oxidoreductase that catalyzes oxidative protein folding in mitochondria. Nat Struct Mol Biol, 16:198–206. 2009.

Banci

, Bertini

, Ciofi-Baffoni

, Boscaro

, Chatzi

, Mikolajczyk

, Tokatlidis

, Winkelmann

. Anamorsin is a [2Fe-2S] cluster-containing substrate of the Mia40-dependent mitochondrial protein trapping machinery. Chem Biol, 18:794–804. 2011.

Bien

, Longen

, Wagener

, Chwalla

, Herrmann

, Riemer

. Mitochondrial disulfide bond formation is driven by intersubunit electron transfer in Erv1 and proofread by glutathione. Mol Cell, 37:516–528. 2010.

Bulleid

, Ellgaard

. Multiple ways to make disulfides. Trends Biochem Sci, 36:485–492. 2011.

Carr

, Winge

. Assembly of cytochrome c oxidase within the mitochondrion. Acc Chem Res, 36:309–316. 2003.

10.

Carrie

, Giraud

, Duncan

, Xu

, Wang

, Huang

, Clifton

, Murcha

, Filipovska

, Rackham

, Vrielink

, Whelan

. Conserved and novel functions for Arabidopsis thaliana MIA40 in assembly of proteins in mitochondria and peroxisomes. J Biol Chem, 285:36138–36148. 2010.

11.

Cavallaro

. Genome-wide analysis of eukaryotic twin CX9C proteins. Mol Biosyst, 6:2459–2470. 2010.

12.

Chacinska

, Koehler

, Milenkovic

, Lithgow

, Pfanner

. Importing mitochondrial proteins: machineries and mechanisms. Cell, 138:628–644. 2009.

13.

Chacinska

, Pfannschmidt

, Wiedemann

, Kozjak

, Sanjuan Szklarz

, Schulze-Specking

, Truscott

, Guiard

, Meisinger

, Pfanner

. Essential role of Mia40 in import and assembly of mitochondrial intermembrane space proteins. EMBO J, 23:3735–3746. 2004.

14.

Dabir

, Leverich

, Kim

, Tsai

, Hirasawa

, Knaff

, Koehler

. A role for cytochrome c and cytochrome c peroxidase in electron shuttling from Erv1. EMBO J, 26:4801–4811. 2007.

15.

Gabriel

, Milenkovic

, Chacinska

, Muller

, Guiard

, Pfanner

, Meisinger

. Novel mitochondrial intermembrane space proteins as substrates of the MIA import pathway. J Mol Biol, 365:612–620. 2007.

16.

Grumbt

, Stroobant

, Terziyska

, Israel

, Hell

. Functional characterization of Mia40p, the central component of the disulfide relay system of the mitochondrial intermembrane space. J Biol Chem, 282:37461–37470. 2007.

17.

Kawano

, Yamano

, Naoe

, Momose

, Terao

, Nishikawa

, Watanabe

, Endo

. Structural basis of yeast Tim40/Mia40 as an oxidative translocator in the mitochondrial intermembrane space. Proc Natl Acad Sci U S A, 106:14403–14407. 2009.

18.

Kloppel

, Suzuki

, Kojer

, Petrungaro

, Longen

, Fiedler

, Keller

, Riemer

. Mia40-dependent oxidation of cysteines in domain I of Ccs1 controls its distribution between mitochondria and the cytosol. Mol Biol Cell, 22:3749–3757. 2011.

19.

Kumar

, Igbaria

, D'Autreaux

, Planson

, Junot

, Godat

, Bachhawat

, Delaunay-Moisan

, Toledano

. Glutathione revisited: a vital function in iron metabolism and ancillary role in thiol-redox control. EMBO J, 30:2044–2056. 2011.

20.

Lionaki

, Aivaliotis

, Pozidis

, Tokatlidis

. The N-terminal shuttle domain of Erv1 determines the affinity for Mia40 and mediates electron transfer to the catalytic Erv1 core in yeast mitochondria. Antioxid Redox Signal, 13:1327–1339. 2010.

21.

Lisowsky

. Dual function of a new nuclear gene for oxidative phosphorylation and vegetative growth in yeast. Mol Gen Genet, 232:58–64. 1992.

22.

Longen

, Bien

, Bihlmaier

, Kloeppel

, Kauff

, Hammermeister

, Westermann

, Herrmann

, Riemer

. Systematic analysis of the twin cx(9)c protein family. J Mol Biol, 393:356–368. 2009.

23.

, Allen

, Wardleworth

, Savory

, Tokatlidis

. Functional TIM10 chaperone assembly is redox-regulated in vivo. J Biol Chem, 279:18952–18958. 2004.

24.

Mesecke

, Bihlmaier

, Grumbt

, Longen

, Terziyska

, Hell

, Herrmann

. The zinc-binding protein Hot13 promotes oxidation of the mitochondrial import receptor Mia40. EMBO Rep, 9:1107–1713. 2008.

25.

Mesecke

, Terziyska

, Kozany

, Baumann

, Neupert

, Hell

, Herrmann

. A disulfide relay system in the intermembrane space of mitochondria that mediates protein import. Cell, 121:1059–1069. 2005.

26.

Milenkovic

, Gabriel

, Guiard

, Schulze-Specking

, Pfanner

, Chacinska

. Biogenesis of the essential Tim9-Tim10 chaperone complex of mitochondria: site-specific recognition of cysteine residues by the intermembrane space receptor Mia40. J Biol Chem, 282:22472–22480. 2007.

27.

Milenkovic

, Ramming

, Muller

, Wenz

, Gebert

, Schulze-Specking

, Stojanovski

, Rospert

, Chacinska

. Identification of the signal directing Tim9 and Tim10 into the intermembrane space of mitochondria. Mol Biol Cell, 20:2530–2539. 2009.

28.

Netz

, Stumpfig

, Dore

, Muhlenhoff

, Pierik

, Lill

. Tah18 transfers electrons to Dre2 in cytosolic iron-sulfur protein biogenesis. Nat Chem Biol, 6:758–765. 2010.

29.

Potting

, Wilmes

, Engmann

, Osman

, Langer

. Regulation of mitochondrial phospholipids by Ups1/PRELI-like proteins depends on proteolysis and Mdm35. EMBO J, 29:2888–2898. 2010.

30.

Reddehase

, Grumbt

, Neupert

, Hell

. The disulfide relay system of mitochondria is required for the biogenesis of mitochondrial Ccs1 and Sod1. J Mol Biol, 385:331–338. 2009.

31.

Sevier

, Kaiser

. Formation and transfer of disulphide bonds in living cells. Nat Rev Mol Cell Biol, 3:836–847. 2002.

32.

Sideris

, Petrakis

, Katrakili

, Mikropoulou

, Gallo

, Ciofi-Baffoni

, Banci

, Bertini

, Tokatlidis

. A novel intermembrane space-targeting signal docks cysteines onto Mia40 during mitochondrial oxidative folding. J Cell Biol, 187:1007–1022. 2009.

33.

Sideris

, Tokatlidis

. Oxidative folding of small Tims is mediated by site-specific docking onto Mia40 in the mitochondrial intermembrane space. Mol Microbiol, 65:1360–1373. 2007.

34.

Stojanovski

, Milenkovic

, Muller

, Gabriel

, Schulze-Specking

, Baker

, Ryan

, Guiard

, Pfanner

, Chacinska

. Mitochondrial protein import: precursor oxidation in a ternary complex with disulfide carrier and sulfhydryl oxidase. J Cell Biol, 183:195–202. 2008.

35.

Terziyska

, Grumbt

, Bien

, Neupert

, Herrmann

, Hell

. The sulfhydryl oxidase Erv1 is a substrate of the Mia40-dependent protein translocation pathway. FEBS Lett, 581:1098–1102. 2007.

36.

Terziyska

, Grumbt

, Kozany

, Hell

. Structural and functional roles of the conserved cysteine residues of the redox-regulated import receptor Mia40 in the intermembrane space of mitochondria. J Biol Chem, 284:1353–1363. 2009.

37.

Terziyska

, Lutz

, Kozany

, Mokranjac

, Mesecke

, Neupert

, Herrmann

, Hell

. Mia40, a novel factor for protein import into the intermembrane space of mitochondria is able to bind metal ions. FEBS Lett, 579:179–184. 2005.

38.

Tienson

, Dabir

, Neal

, Loo

, Hasson

, Boontheung

, Kim

, Loo

, Koehler

. Reconstitution of the mia40-erv1 oxidative folding pathway for the small tim proteins. Mol Biol Cell, 20:3481–3490. 2009.

39.

van der Laan

, Bohnert

, Wiedemann

, Pfanner

. Role of MINOS in mitochondrial membrane architecture and biogenesis. Trends Cell Biol, 22:185–192. 2012.

40.

, Dailey

, Wang

, Rose

. The crystal structure of augmenter of liver regeneration: A mammalian FAD-dependent sulfhydryl oxidase. Protein Sci, 12:1109–1118. 2003.

41.

Zapun

, Missiakas

, Raina

, Creighton

. Structural and functional characterization of DsbC, a protein involved in disulfide bond formation in Escherichia coli. Biochemistry, 34:5075–5089. 1995.

42.

Zhang

, Lyver

, Nakamaru-Ogiso

, Yoon

, Amutha

, Lee

, Bi

, Ohnishi

, Daldal

, Pain

, Dancis

. Dre2, a conserved eukaryotic Fe/S cluster protein, functions in cytosolic Fe/S protein biogenesis. Mol Cell Biol, 28:5569–5582. 2008.